School Closures During the 1918 Flu Pandemic

Abstract

The COVID-19 pandemic has reignited interest in responses to the 1918-19 influenza pandemic, the last comparable U.S. public health emergency. During both pandemics, many state and local governments made the controversial decision to close schools. We study the short-and long-run effects of 1918-19 pandemic-related school closures on children. We find precise null effects of school closures in 1918 on school attendance in 1919-20 using newly collected data on the exact timing of school closures for 168 cities in 1918-19. Linking affected children to their adult outcomes in the 1940 census, we also find precise null effects of school closures on adult educational attainment, wage income, non-wage income, and hours worked in 1940. Our results are not inconsistent with an emerging literature that finds negative short-run effects of COVID-19-related school closures on learning. The situation in 1918 was starkly different from today: (1) schools closed in 1918 for many fewer days on average, (2) the 1918 virus was much deadlier to young adults and children, boosting absenteeism even in schools that stayed open, and (3) the lack of effective remote learning platforms in 1918 may have reduced the scope for school closures to increase socioeconomic inequality.

Lifetime Consequences of Lost Instructional Time in the Classroom: Evidence from Shortened School Years

“…They found corresponding reductions in labor earnings of 3.2% and 1.9% measured at ages 30-40. In contrast, Ager et al (2020) find no effects of school closures during the 1918 flu pandemic in the U.S. on wages measured at slightly younger ages. They argue, however, that school attendance and educational attainment did not respond to local variation in school closures and reopenings because many people stayed home independent of local policies.…”

“…What are the long-run consequences of lost instructional time in the classroom? This question has important policy implications in the context of various situations that force students to temporarily stay away from school such as inclement weather conditions (e.g., Marcotte and Hemelt, 2008;Goodman, 2014), natural disasters (e.g., Sacerdote, 2012), teacher strikes (e.g., Belot and Webbink, 2010;Baker, 2013;Jaume and Willén, 2019), prolonged summer holidays (e.g., Kuhfeld et al, 2020), or the spread of infectious diseases (e.g., Ager et al, 2020;Meyers and Thomasson, 2021). The latter case has recently gained prominence worldwide due to the outbreak of the COVID-19 pandemic, which led to school closures affecting millions of students around the globe (UNESCO, 2021).…”

To What Extent Does In-Person Schooling Contribute to the Spread of COVID-19? Evidence from Michigan and Washington

“…Descriptive studies of nonpharmaceutical interventions during the 1918 flu pandemic found that cities that closed schools had lower death rates than cities that did not close schools and that cities that implemented control measures that included school closure had lower spread (Bootsma & Ferguson, 2007;Winslow & Rogers, 1920). School closures during the 1918-1919 flu pandemic did not appear to have negative effects on student learning or future adult outcomes such as wages-though these closures lasted for shorter periods than COVID closures (Ager et al, 2020).…”

Do Class Closures Affect Students' Achievements? Heterogeneous Effects of Students' Socioeconomic Backgrounds

“…Since the school closures due to the pandemic of COVID-19 were nationwide, however, all of these studies rely on the comparison of before and after the COVID-19 pandemic as their identification strategy. Another previous studies analyzed the effects of school closures using a variation of instruction time reduced by class closures, based on historical pandemics such as the 1918-1919 influenza and the 1916 polio epidemic in the United States (Ager et al, 2020;Meyers and Thomasson, 2021). But these studies focused on factors that were not directly related to the timing of school closures, such as the years of schooling and labor market outcomes.…”